From Gross Anatomy to Nanotechnology

Behind the Science: A conversation with Ulrich von Andrian

Ulrich von Andrian. Image: John Soares

Ulrich von Andrian, the Edward Mallinckrodt Jr. Professor of Immunopathology in the Department of Microbiology and Immunobiology, studies how white blood cells migrate around the body and interact during an immune response.

But von Andrian wasn’t always involved in immunology. In this conversation he traces his journey from an institute in Germany where the labs looked like the set for “Star Trek” and the lab in La Jolla where graduate students were, to his surprise, both seen and heard.

He recalls when cloning was new and now looks to nanotechnology as the next tool for instructing the immune system. His fascination with scientific research grabbed him early, and hasn't let go.

HMS: What drove you to become a physician-scientist?

VON ANDRIAN:  I came late to both medicine and science, and in that order. Growing up in Munich, Germany, I would say I was not an excellent student. I took a more minimalist approach to academic pursuits. My interest was more in playing soccer and tennis, having fun as a kid. But I always felt intrigued by science classes.

I grew up without a father. My father passed away from cancer when I was just 11 years old. My mother was working as a technician in a medical laboratory, so I had a little bit of exposure at home to the idea that there's lab work to be done in the context of medicine.

I then went to medical school, but in Germany, if you graduate from medical school, you get a degree as a physician but you're actually not a doctor until you do a medical thesis.

I had a very influential godfather, who was a vascular surgeon in Munich and who told me about the Institute for Surgical Research, which in the early 1980s was a very famous place. My godfather told me that's where I should go and do my thesis.

HMS: What was it like?

VON ANDRIAN:  It was like walking on the set of “Star Trek.” There were operating rooms with fancy things they called computers, which were not really commonplace at the time. That was such a wow experience for me.  I thought to myself, I want to do something really amazing here.

HMS: What drew you to immunology?

VON ANDRIAN:  In gross anatomy, you learn where everything is located. Ultimately the science of anatomy is the science of localization, but one organ that is not definable based on its localization is the immune system. This makes perfect sense because the immune system has evolved to deal with infections that can occur anywhere in the body. You have to rapidly shift forces to deploy them wherever they are needed.

This has always been a fascination of mine. My early career was trying to understand how leukocytes in the bloodstream—white blood cells—can get recruited into tissues, how you make this process efficient, how you make it specific, and how it may explain some of the unique features you see not only in response to infections but also in inflammatory and autoimmune disease. You can have highly selective organ involvement and highly specific subsets of immune cells that get recruited to a given tissue while others remain excluded from that same tissue.

HMS: How do you study the immune response?

VON ANDRIAN: You can’t really discover how these things work by just going to a petri dish. You would just get a weak echo of what really goes on in a living animal. We are trying to study things that occur in a highly dynamic system where individual cells have to make decisions in fractions of seconds as they go at extremely high speed through blood vessels.

So the only way to begin to comprehend these processes is by actually watching them happen in a living organism in real time. My lab is in a long tradition of laboratories that uses intravital microscopy to directly image immune cell migration.

HMS: How did you get from medical training to basic research? 

VON ANDRIAN: There’s a fairly serendipitous, but nonetheless straight, line from my first experience in Munich at the Institute for Surgical Research, when I felt I wanted to be a neurosurgeon. The lab I was in was trying to address a problem that happens after a brain injury, where blood vessels in a damaged region often become leaky, so water leaks out into the brain. It's called brain edema.

One of the hypotheses that I tested was in rats. If there's tissue injury, that may induce an inflammatory response. And because of the inflammatory response, you may get neutrophils—a type of white blood cell that helps fight off infections—accumulating in the brain. Neutrophils at the time were understood to release enzymes that can degrade tissues, like elastase.

So I tested one of the very first recombinant proteins that were given in significant amounts, an elastase inhibitor called eglin-c, which had been cloned out of the salivary gland of leeches. And it did absolutely nothing in my model. But it got me interested in neutrophils.

HMS: From the experimental surgery institute in Germany you landed in California.

VON ANDRIAN: I went to the lab of Karl Arfors, who at the time had a laboratory affiliated with UCSD in La Jolla.

I owe a lot to him. He was my supervisor, but he also was really a friend, and a very supportive friend. The system I grew up in, in Germany, was much more traditional, much more hierarchical. As a graduate student, you were sort of at the lower part of the totem pole.

After I first started in Karl's lab, I wasn't even there for a week when I was just listening to him and another more senior member of the lab discussing some scientific question. I had a thought that made me sort of make a noise, and they both stopped what they were saying and looked at me and really wanted my opinion. I wasn't used to voicing opinions.

I think that is a very important lesson I learned on that occasion, which is that everyone who has a head on their shoulders can think with it regardless of what their job title is.

HMS: What did you work on in Arfors’s lab?

VON ANDRIAN: When I arrived, there was a very strong interest there in adhesion molecules, and we initiated a collaboration with Eugene Butcher at Stanford, where I would later do my second postdoc. 

In order to actually get neutrophils to go to a tissue and cause damage, they first need to adhere to endothelial cells to be able to leave the bloodstream. And it was kind of confusing because there were so many, and it was unclear why you need so many. They all seemed to be doing the same thing: conferring some mechanical stability to interactions between cells.

There's actually a sequence of molecular interactions, and these adhesion molecules can be categorized based on when they take action in that chain of events. There are some that are specialized and engage in interactions under very high-shear conditions forming some very rapid bonds and they form this rolling type interaction.

But then there's a second set of adhesion molecules that are actually under control of additional stimuli that the cells must receive in order to be turned on. So they're in a resting state until they're activated. So while the cell is rolling, it needs to receive an activating stimulus that turns on the second set of adhesion molecules. And when those get activated, now the cell can bind and throw the anchor.

So we published; that was my first paper from that time. And is now in all the textbooks.

HMS: What has the impact been?

VON ANDRIAN: It turns out to be a broad principle that really allows you to understand how different types of leukocytes get selectively recruited to different tissues, and why different tissues have different preferences, and different diseases have different preferences for what kind of cells they're recruiting.

And so, the counter aspect to that is that if there are unique multistep combinations that basically function like area codes—you have a certain sequence of molecules that's very unique to a given tissue, a different type of pathologic cell—you understand which is the bottleneck there, you could make a drug that targets that. It will more or less selectively interfere with the pathologic inflammatory event, but it won't paralyze the entire immune system.

HMS: What else did you learn from your mentor?

VON ANDRIAN: Karl, although Swedish, speaks German pretty well. He would repeat a German saying: probieren ist besser als studieren, “to try is better than to study.” Do a lot of experiments. Acquire data. Try to understand the data and see what you can learn from this, see what comes out on the other end.

Some of the things I did during my early days as a postdoc seemed rather desperate. Now I think they were some of the most valuable experiences I had, because it made me think about so many different things related to microvascular phenomena. I was forced to have a very open mind looking through the microscope every day, and all the little things that you could see happening there.

HMS: Was there a pivotal moment?

VON ANDRIAN: There was a critical event that I think completely changed my trajectory, which was my first public talk in English, other than at a lab meeting. That was when I had done intravital microscopy studies demonstrating the sequence of events in the leukocyte adhesion field.

There was an invitation-only meeting organized by Tim Springer at Cold Spring Harbor Laboratory, one of the so-called Banbury Center conferences.

Eugene Butcher was supposed to be there and Karl was also invited, but he let me go and take his spot. The first evening before the first day of the three-day conference I saw Tim Springer for the first time. I had known him from the literature so it was very nice to talk to him. Now he's a good friend. Back then he said,

            ‘So what's the title of your talk?’

            ‘What talk?’

            ‘Well, Eugene Butcher has sent a fax that he wasn't coming,

                  that you were talking.’

             So I said, ‘I didn't know that.’

             ‘Oh, you're not going to talk?’

             ‘Now I'm going to talk.’

 So this talk was scheduled for the second day. Remember, there was no email then. I had the folks in La Jolla FedEx me a few slides and videotapes of my experiments. And then I spent the whole day walking through this beautiful park in the middle of the night, talking to the trees, practicing that talk.

I was fortunate because Tim Springer gave a lecture the day before. And his lab in parallel—I had no idea—had basically recreated this multistep adhesion cascade in vitro. So he gave a talk, and the audience was very intrigued, but also puzzled. It wasn't really clear to anyone what to make of it. And I got up the next day.

No one knew me; I had never given a talk in English before. I said, "I'm from Karl Arfors's lab, and I do intravital microscopy, and actually Tim Springer has given the perfect introduction.”

And so, I present—speak about luck! It was a half-hour talk, and I had only like six or seven figures from my paper, which was published in PNAS. I didn't have a choice: I just used the blackboard and a piece of chalk to draw all these things.

There was a tremendous excitement. What we had learned about leukocyte adhesion was a real revelation to them and just a fantastic feeling for me. Within a half hour, I hadn't even published anything yet in this field, but I was basically one of the players in the field. And have been ever since.

I had a much more heavy German accent then than now. Actually, I was constantly told that I sounded like Arnold Schwarzenegger. This was the time when “The Terminator” had just come out.

HMS: What was it like to create recombinant proteins when cloning was new?

VON ANDRIAN: In those days, cloning was all the rage. You couldn’t open an issue of Nature, Science or Cell without another adhesion molecule that was cloned.

Learning how to design these chimeric molecules was really fun. I learned how to do sequence overlap extension PCR—polymerase chain reaction—it was very interesting.

I made expression vectors, put them into cell lines and tried to stain them with antibodies. And the antibodies actually stained the cell, so it worked; something was expressed. I looked through the microscope and it was just fluorescent cells. Big deal, right?

But I remember looking through the microscope at fluorescent cells expressing a protein that I made, that hadn't been conceived by nature. It's this sort of divine feeling, you know? This amazing sense of being able to harness the principles of nature to do things that come out of your mind. It was just spectacularly cool.

HMS: What would people be surprised to learn about you?

VON ANDRIAN: I never in my life had a single lesson in real immunology. I was trained as a physician to have a reasonable understanding of how the mammalian body works, how it all hangs together, how one part influences the other parts. But many PhDs I see today, they get trained to really have very deep knowledge about a rather limited area. You need to be able to have the big picture, but then you also need to be able to dig deep and have very high resolution in certain areas.

To me, it was very helpful to have training that emphasized the big picture first. And once I'd committed to specific questions that I wanted to address in biology, I would work on understanding the details in those areas, but in the context of the whole. That, to me, makes it easier now. I always like to have medically trained members in my lab in addition to PhDs because I think the two can really learn a lot from each other because of the very different training they have.

HMS: What do you think about the scientific enterprise as it currently exists?

VON ANDRIAN: To me, one of the biggest challenges we face is how to translate scientific insights into medicine. It's a structural and organizational and financial challenge. I think the current system is very ineffective. And I think it has to do in part with the demands that are placed on making a scientific discovery and taking that scientific discovery and driving it through a process where ultimately you may end up with a new treatment. The expertise that is needed to make the discovery and to actually develop that discovery is enormously different.

And I think there are many challenges in really effectively communicating the essentials between the many different stakeholders that are sort of lined up along the path. I don't have a solution, but I witness this often enough to say that there should be a better way.

HMS: You have entered the fray with a biotech company called Selecta Biosciences.

VON ANDRIAN: I did cofound a company in Watertown with Omid Farokhzad of Brigham and Women’s and Bob Langer from MIT that is making nanoparticles to communicate with the immune system. One way you can use these nanoparticles is to make vaccines or to deliver an antigen in a particular context.

Modern immunology teaches us there are a virtually unlimited number of molecular structures that are foreign to our body. All of these could potentially be antigens. If I drink my coffee, there is protein and all manner of stuff in there. I don’t want to have an immune response to my coffee, although it’s all foreign.

So recognition of a molecular structure as an antigen per se is not sufficient to make an educated immune response possible. There always has to be context. There are a variety of pathways that need to be engaged in parallel or in a very precisely timed sequence so you get a desired or even an undesired immune response.

I think nanotechnology is a fantastic way to present contextual instructions to the immune system. You can make synthetic particles that are the size of a virus, so if you give them to the body they already have the shape and appearance of what the immune system has evolved to recognize. You can equip them with a multitude of signals that together form precise instructions.

In traditional pharmacology we are used to making drugs that are small molecules or even antibodies that interfere with just one pathway at a time. So you can interfere with certain signaling events by going after some bottleneck more or less well, but if you want to provide instructions to the immune system to get a desired response, that usually doesn’t work well. It’s as if I was talking to you in single words. I need to use sentences for the grammar where you have different symbols in a precise sequence. Otherwise I can’t make myself understood.

HMS: What is your approach to running a lab?

VON ANDRIAN: I like to recruit postdocs who have some relevant experience, but who have different backgrounds, who bring with them techniques, knowledge and understanding that no one in the lab already has. Because I haven't had formal training in many of the things that are now sort of daily activity in the lab, I need to learn these things from my lab members. So it's good to bring on people who have done really well elsewhere in areas that I feel I don't understand enough, to learn from them.

Whoever comes to me gets a lot of freedom and independence, usually. I'm hopeless at micromanaging. I will not even try that. I make suggestions and I will give advice and I will voice opinions, but, otherwise, if someone has a good idea and it's not my idea, that's fine.

Also, to be successful, of course you have to work hard. But whoever comes to my lab, one of the first things I tell them is that they're here because of their career, not because of my career. And there is no good correlation between the hours anyone spends in the lab and the quality of the work that is being done. I can only help them to do good work, but I can't do good work for them. And for many people of whom I'm very proud, this has worked out very well, and for some it hasn't.